SUMMARY Household air pollution (HAP) is the third leading risk factor for morbidity and mortality worldwide, and the leading cause of respiratory disease in low- and middle-income countries (LMICs). HAP is largely the result of inefficient combustion of solid fuels (biomass) for cooking and heating. Biomass fuel is the main domestic energy source used by ~50% of urban households and ~90% of rural households globally and is a potential public health intervention with significant impact on morbidity and mortality. Although there is evidence that reducing HAP results in better health outcomes, few interventions aimed at more efficient or better ventilated cook stoves which utilize biomass (i.e clean cook stoves) have produced significantly improved pulmonary health outcomes. This is largely due to the fact that prior interventions have not use clean fuels and could not reduce HAP to levels the produce meaningful outcomes. Liquefied Petroleum Gas (LGP) has been proposed as a clean fuel alternative to traditional biomass cook stoves. We hypothesize that reducing HAP with LPG intervention will produce improved pulmonary health outcomes among adults and that these improvements will be dose-dependent. To study the effect of HAP on pulmonary health, we propose to apply detailed respiratory questionnaires and conduct spirometry among a subgroup (n=400) of adult women enrolled in an existing study, The Household Air Pollution Intervention Network trial (HAPIN). This proposal will leverage extensive environmental exposure data collected as part of the parent grant to assess the exposure-response relationship between HAP and respiratory symptoms and lung function. This proposal has two novel aims. First, to determine whether LPG intervention results in improved respiratory symptoms and improved lung function, we will apply detailed respiratory questionnaires and conduct spirometry at three time periods in the trial. Second to assess whether this effect is responsive to the type and amount of exposure, we will utilize environmental data assessing P2.5, carbon monoxide and black carbon exposure using novel causal inference measures. Results from this study will inform public health interventions aimed at reducing HAP and the burden of disease globally.